Staff / Senior Machine Learning Engineer

About DKatalisDKatalis is a financial technology company with multiple offices in the APAC region. In our quest to build a better financial world, one of our key goals is to create an ecosystem-linked financial services business.DKatalis is built and backed by experienced and successful entrepreneurs, bankers, and investors in Singapore and Indonesia who have more than 30 years of financial domain experience and are from top-tier schools like Stanford, Cambridge London Business School, JNU with more than 30 years of building financial services/banking experience from Bank BTPN, Danamon, Citibank, McKinsey & Co, Northstar, Farallon Capital, and HSBC.About the roleWe are hiring a Staff Machine Learning Engineer to help us turn data science work into a reliable, customer-facing product, and we will also consider strong Senior candidates. This is a hands-on senior IC role. At Staff level, you will set technical direction for how we ship and operate ML in production; at Senior level, you will own significant slices of that work and help shape direction alongside the team.About the team and the workYou'll join our Data Science team to build the ML systems powering a digital bank serving millions of customers. Current work spans:A broad portfolio of production models across fraud, collections, scoring, and customer-facing features, with active investment in shared monitoring, retraining, and feature infrastructure.Real-time credit and risk decisioning in the loan origination flow.A multi-cloud LLM platform under active build, supporting LLM-backed product features such as intelligent document understanding, agentic assistants, and conversational banking.The constraints are real (regulated environment, customer money, data residency requirements), and the platform is still being shaped, so there is genuine build work to do, not just maintenance.Key Responsibilities:Own and evolve our full-stack ML platform (the unified system through which every model is built, served, monitored, and retrained) so that data scientists ship business value faster and more reliably.Compress time-to-production for new models, turning research into customer-facing features in weeks rather than quarters, with MLOps practices that make each launch routine rather than bespoke.Protect the business from model risk by using the platform's drift detection, model monitoring, and A/B and canary deployment capabilities to catch degradations before they affect customers or revenue.Unlock new product surfaces (credit decisioning, fraud, personalization, LLM-backed experiences) by offering real-time and batch inference with the latency, reliability, and feature store consistency they require.Reduce the cost and effort of running ML in production by using standardized CI/CD for ML, model versioning, reproducible training, and clear operational ownership.Mentor engineers, partner with data, product, and platform teams, and set the standards that shape how DKatalis does ML.Your work will support various business capabilities, including:Digital banking and financial product features (e.g., smart financial recommendations, personalization)Credit and risk scoring (real-time and batch)Fraud detection and risk managementGrowth, go-to-market, and customer engagement strategiesImproving the efficiency of technical and business operationsRequirements:Technical Skills and ExperienceWe expect strong software engineering and ML foundations, plus depth in one of two MLE specializations. If you have most of the following, please apply — we don't expect any single candidate to tick every box.5+ years (Senior) or 8+ years (Staff) of software engineering, with substantial time shipping production machine learning.Production-grade Python and software engineering practice: clean code, typing, testing (pytest), code review, and systems design for end-to-end ML pipelines (batch and streaming, capacity and latency planning, API design).Cloud and infrastructure fluency on GCP (Vertex AI, BigQuery, GKE, Cloud Run, Pub/Sub) or AWS (primarily SageMaker and Bedrock). We mostly use GCP. Comfort with Kubernetes and infrastructure-as-code (Terraform) is required.Strong SQL and data warehouse skills, with attention to query and storage optimization (BigQuery in particular).Depth in one of the two MLE directions:Applied ML and algorithms: model selection, evaluation, feature engineering, and familiarity with some of the PyData stack (e.g., pandas, scikit-learn, PyTorch, or TensorFlow).MLOps and platform / SRE: workflow orchestration (e.g., Kubeflow, Vertex Pipelines, Airflow), CI/CD for ML, observability, and SLOs.Production ML ExpertiseDepth in whichever direction matches your specialization:Applied ML and algorithms:ML fundamentals: supervised learning algorithms, evaluation metrics, validation strategy, overfitting, and judgment on when deep learning is the right tool.Feature engineering and handling data quality, missing data, and drift.Designing offline and online experiments: A/B tests, holdouts, choosing success metrics.MLOps and platform / SRE:Model monitoring: performance, data drift, concept drift, silent failures, alerting, and SLOs.Deployment patterns: canary, shadow, blue/green; model versioning and rollback.Feature stores and the realities of feature freshness, lineage, and training-serving skew.Hands-on experience shipping LLM-backed product features (RAG, prompt management with Langfuse or equivalent, offline and online evaluation, guardrails, cost and latency management) is a plus.For Staff CandidatesIn addition to the above, we expect Staff candidates to demonstrate:Working fluency in the second MLE direction (a strong MLOps engineer with a solid grounding in applied ML, or vice versa) and breadth across the full production ML lifecycle.A track record of setting technical direction across multiple teams or workstreams.Driving platform-level improvements that compound across projects (e.g., shared serving infra, monitoring frameworks, deployment templates).Cross-functional leadership: running incident response, defining standards, and influencing roadmaps with data and product partners.Mentoring and growing engineers and data scientists in production ML practices.Soft SkillsExcellent communication skills, both written and verbal; able to explain technical trade-offs to product, business, and non-technical audiences.Strong ownership: comfortable being the person ultimately accountable for the systems you build.Pragmatic and outcome-oriented; able to handle ambiguity and changing priorities.Intellectual humility: willing to admit knowledge gaps and ask for help.Nice to HaveExperience with dbt or similar SQL transformation tooling.Hands-on experience with real-time serving stacks (Kafka, Redis, low-latency online inference).Experience in retail banking, fintech, or other regulated sectors.On-call and incident management experience for ML systems.Proven track record of improving ML infrastructure efficiency, reliability, or cost.Contributions to open-source ML, MLOps, or data engineering projects.Strong portfolio of relevant projects (e.g., well-maintained GitHub repositories), public writing or talks, or other community involvement.